Monday, April 25, 2011

Small Group Size Limits and Self-Reinforcing Feedback Loops

In every revolutionary organization I've been a part of, one thing has been constant -- the number of people consistently coming to local meetings was roughly 5-15, though generally a more limited 5-8 people.  We would try to recruit new people, make the work of the organization more relevant, or find ways to entice inactive members back, but any gains were short-lived.

There is a reason for this.  It seems that at least part of this is due to fundamental limits to organizations.  From evolutionary anthropology, sociology, evangelical church strategy, business research, and statistics, some common threads keep coming up.

In "How to Design Small Decision Making Groups" a dilemma is discussed.  The more people you have in a group, the more likely someone will propose the correct solution to a problem.  But the more people you have, the more communication is needed.  In fact, they show that "the difficulty of managing communication is roughly proportional to the number of possible social interactions" and that the number of possible social interactions explodes exponentially for groups larger than 5.  If you consider all possibilities of subgroup communcation, there are only 26 possibilties for a group of 5.  This grows to over 1000 for a group of 10 and over 1,000,000 for a group of 20.

I've been in many meetings of 5 or fewer people where people grew frustrated or things took too long, but this was generally due either to the group not cutting off a member's long monologues or to a fundamental disagreement where no amount of argument would lead to a consensus.  These are real issues that can be addressed with some basic ground rules and care.

But for larger groups, especially as you approach 10 or even larger, I've seen a greater sense of frustration, where everyone wants to speak and no-one feels like they get a chance.  In these cases, the problem is generally a group process that doesn't scale well combined with basic math.

Imagine a simple scenario where all N people in a group are given the opportunity to speak for 2 minutes, and everyone can respond to others people's initial point for 1 minute.  The meeting lasts 2N+N(N-1) minutes which is 30 minutes for 5 people, almost 2 hours for 10 people, and 7 hours for 20 people.  The problem is clear here, even with severe limitations on communication.  Allowing fuller interaction would make the meeting of 5 take maybe another hour, but 20 people might take literally years of nonstop 24/7 meetings.

So why don't we see meetings of 20 people running on indefinitely forever?  One answer is that some groups find formal structures and processes that limit communication and still hold onto both democracy and effectiveness.  But while these cases raise the reasonable limit to group size, examples where it is raised to 20 or 30 are few and far between.

For most, even in groups that successfully raise the limit with formal processes, there's another answer.  Long before the problem gets impossibly bad it gets frustrating and boring.  Everyone has a different threshold.  Stubborn true-believers with nothing better to do with their time have a high threshold, and busy single parents with two jobs and a skepticism that maybe this group isn't actually going to change the world have much lower thresholds.  As people hit their threshold, they either quit in frustration or gradually disengage as they find more and more higher priority demands on their time.

The result is that as the group grows, it reaches a point where meetings are frustrating or useless to some of the members with low thresholds, and those members leave or become inactive.  If this happens suddenly it can cause a crisis in the group, as the group tries to find a solution to the problem with attempts to get everyone to express their thoughts.  As you can imagine, if the communication explosion was the source of the problem, this causes things to spiral out of control until either a large number of people have left or the group finds another way to drastically limit communication.

But without a crisis, another interesting thing can happen.  As long as a bare minimum level of recruitment is maintained, the group reaches a steady state and stays at a nearly constant size regardless of any additional rate of recruitment.  In fact, any attempt to grow the group through increased recruitment not only paradoxically fails to grow the group, but it has a primarily negative impact - leading to increased turnover, a lower proportion of experienced members, and less diversity due to a greater proportion of people with characteristics associated with higher thresholds, as well as a diversion from other activities of the group.  Again, attempts to address this head-on by quickly scaling up recruitment until the group's growth starts to take off can cause things to spiral out of control.

While less than sustainable recruitment leads to the slow death of a group, this self-correcting feedback loop gives us direct experience that recruitment above that level is at best pointless.  If minimal sustainable recruitment can be achieved without specific thought or effort, then it becomes difficult to see any value in recruitment at all, especially since an organization that is doing good work should be able to attract a minimal number of recruits without additional effort.

For some pro-organizational revolutionaries, this isn't a problem.  Whether it's a study or propaganda group, a cadre group, a project-based group, or even a direct action group, many people have found that a small group is fine.  With a low level of recruitment, the organization can become tight as a small number of people with a high level of agreement and the same high level of involvement learn and grow together.

But that isn't the only option.  For those that seek to build a growing revolutionary organization, this limiting feedback loop presents a dilemma.  With even a low rate of recruitment, a revolutionary organization should be able to grow from 10 to 50 active members in a single city within a few years, opening up new possibilities for action.  Yet this sort of growth is practically unheard of.

I've talked a bit about why it seems like there is a barrier around 10 members that groups frequently fail to penetrate.  In fact, according to "Discrete Hierarchical Organization of Social Group Sizes", evidence suggests there is a series of barriers, with roughly each tripling in group size requiring fundamentally different structural and relational characteristics.  This idea can help us to imagine how a revolutionary organization might grow, crossing these barriers.

Next: Beyond Small Groups - Organizational Growth and Phase Changes

1 comment:

  1. This is some very valuable stuff Greg. Keep it coming!